The present invention relates generally to information recording and/or reproducing systems, and more particularly to improved means and methods for positioning a transducer relative to a moving medium.
The following United States patents are of interest with respect to transducer positioning systems:
U.s. pat. No. 3,491,347 to Farrand PA1 U.s. pat. No. 3,691,543 to Mueller PA1 U.s. pat. No. 3,686,649 to Behr PA1 U.s. pat. No. 3,699,555 to Du Vall PA1 U.s. pat. No. 3,812,533 to Kimura et al. and PA1 U.s. pat. No. 3,838,457 to Palmer.
The following article is also of interest: "Design of a Disk File Head-Positioning Servo," R. K. Oswald, IBM Journal of Research and Development, November 1974.
Reference is also made to the related commonly assigned patent application Ser. No. 572,303, filed Apr. 28, 1975, for TRANSDUCER POSITIONING SYSTEM, R. S. Kril, inventor.
A common type of device to which the present invention may be applied is a magnetic disc recording and reproducing system employing one or more read/write heads as transducers. Such a system is typically employed for providing auxiliary storage in a data processing system. In such a system, data is recorded on a disc in a plurality of concentric circumferential tracks. Data located in a particular track of a disc is read by properly positioning the read/write head in a radial direction so as to be located directly over the desired track. The radial positioning operation of the head typically involves two steps: (1) a wide range coarse positioning step (also referred to as track seeking) for moving the head to a radial position within the immediate proximity of a desired target track, and (2) a fine positioning step for bringing the head into accurate alignment with the target track and for maintaining this alignment (also referred to as track following) until it is desired that the transducer be repositioned to a new track.
As will be evident from the aforementioned patents and article (e.g. see U.S. Pat. Nos. 3,691,543; 3,812,533; and 3,838,457 and the aforementioned article), fine positioning is usually achieved by controlling the head positioning carriage in response to the detection of pre-recorded encoded servo data using either the same head as is used to read the work data, or a special servo head. The servo data may be recorded on the same disc as the work data or on a separate disc or surface having a precise mechanical relationship thereto. Coarse positioning is typically achieved in two common ways: (1) by controlling radial head movement based on detection of the movement of the head positioning carriage, such as by employing photoelectric detection means (e.g. see U.S. Pat. No. 3,812,533), or (2) by controlling radial head movement based on detecting track crossings using the recorded servo data provided for fine positioning purposes (e.g. see U.S. Pat. Nos. 3,691,543, and 3,838,457).
The latter of the above two coarse positioning approaches employing the fine positioning servo data for coarse positioning detection is to be preferred in many applications, since it does not rely on establishing and maintaining a precise relationship between movement of the carriage and the head. However, a problem arises in providing coarse positioning in this preferred manner in systems in which the servo data is interspersed on the same disc as the work data, such as where the work data and the servo data are arranged in alternating sectors on the disc, as is commonly done. The problem arises because of the possibility of the head crossing a plurality of tracks between servo sectors during coarse positioning, which is normally to be expected in high density systems. In order to accurately count track crossings in such a situation, known systems have had to expand the capabilities of the recorded fine positioning servo data to permit uniquely identifying a group of adjacent tracks rather than just two tracks, the number of tracks in the group being at least equal to the maximum number of tracks which may be crossed by a head between servo sectors. While this expansion of the recorded fine positioning servo data solves the problem, it has the significant disadvantage of requiring a relatively large portion of the disc surface for its implementation because of the need to maintain the integrity of its fine positioning capabilities in the expanded form.
Even in those applications in which servo data is recorded on a separate disc or on a separate disc surface (e.g. a dual layer disc), it is still desirable to provide for uniquely identifying a group of tracks to permit electrically off-setting the servo head as may be required to bring the work data head into exact registration with a work data track. The need for this capability occurs, for example, when one device is used to record data on a disc, and the disc is then physically removed and stored for eventual mounting on a second device which might provide a slightly different spacing between the work data and servo data heads than existed on the first device. As pointed out above, the expansion of the conventional recorded fine positioning servo data to permit unique identification of a plurality of tracks would likewise require a relatively large track storage area.